An adaptationist framework predicts that people's intuitions regarding modern criminal justice stem from psychological mechanisms designed for countering the evolutionarily recurrent equivalent, exploitation (Petersen et al. Reference Petersen, Sell, Tooby, Cosmides and Høgh-Olesen2010). By implication, the model of evolved counter-exploitation mechanisms described by McCullough et al. should predict core features of criminal justice intuitions.
Most of the previous applications of the adaptionist framework to criminal justice focus on people's perceptions of crime seriousness, arguing that they emerge from evolved intuitions about harm (e.g., Robinson et al. Reference Robinson, Kurzban and Jones2007). In making this argument, researchers emphasize the high levels of consensus among individuals and countries on the relative seriousness of different crimes. If widespread agreement is a necessary product of the operations of evolved mechanisms, we would conclude that such mechanisms play little role in the production of the intuitions most relevant to McCullough et al.'s model: that is, intuitions about whether punitive or reparative goals should be given priority in criminal justice. Hence, in criminological research, consensus that people have incoherent or “mushy” criminal justice intuitions has emerged from observations of profound disagreement between individuals, and in different situations about the prioritization of punitive and reparative sanctioning goals (Cullen et al. Reference Cullen, Fisher and Applegate2000; Roberts & Hough Reference Roberts and Hough2002; Roberts & Stalans Reference Roberts and Stalans2004).
McCullough et al.'s model emphasizes (1) the existence of computational systems designed to implement revengeful as well as forgiving strategies, and (2) computations of the social value of the exploiter as the key regulator of the activation of these strategies (see also Petersen et al. Reference Petersen, Sell, Tooby, Cosmides and Høgh-Olesen2010). These two features, I argue, render disagreement between individuals and in different situations an unavoidable outcome of the operation of our evolved counter-exploitation intuitions.
This relates to the fact that the social value of others is both self-specific and target-specific. One target is the self's kin, another is not; in one case, the target and the self share a history of cooperative interactions, in another case they don't; one target is part of the same group as the self, another is not. And so on. Accordingly, different selves will compute the social value of the same target differently, and a single self will compute the social value of different targets differently. In the face of a specific exploiter, some individuals will therefore experience punitive sentiments, while others will experience reparative sentiments.
In modern mass societies, an individual's personal welfare will rarely be affected by how the state sanctions a single criminal, regardless of that criminal's social value. Nevertheless, given the evolutionary importance hereof, our minds should be designed to automatically pick up on ecologically valid cues to the presence of social value (e.g., expressions of remorse, shared ethnic background, lack of criminal record) and activate punitive and reparative sentiments accordingly. In modern criminal justice, different sanctioning types are aligned with these different sentiments to varying degrees. Cost-imposing sentences such as prison are aligned with punitive sentiments, whereas rehabilitation schemes are more aligned with reparative sentiments (e.g., Cullen & Gendreau Reference Cullen and Gendreau2000). Accordingly, different individuals will prioritize different sanctioning types in the face of the same criminal (e.g., depending on the match between the respective ethnic or racial backgrounds of the self and the criminal), and the same individual's priorities will change in the face of different criminals (depending on the specific cues surrounding each case).
This simple observation readily integrates the adaptationist framework and the demonstrations of disagreement within individuals towards different crimes and between individuals towards the same crime. According to this interpretation, lay intuitions are not “mushy,” and disagreement is driven by the existence rather than lack of principled intuitions. If valid, we should find that whether individuals agree or disagree about appropriate sanctions is predictable from whether there is agreement or disagreement with respect to the perceived future social value of the criminal. Some evidence for this link already exists (Burnette et al. Reference Burnette, McCullough, Van Tongeren and Davis2012; Lieberman & Linke Reference Lieberman and Linke2007; Petersen et al. Reference Petersen, Sell, Tooby and Cosmides2012).
Modern criminal justice intuitions have been deemed “mushy” for more than just these kinds of disagreement. In particular contexts, research has also demonstrated how people find it highly difficult to prioritize either reparative or punitive sanctioning goals. Instead, people concurrently express firm support for both (Cullen et al. Reference Cullen, Fisher and Applegate2000; McCorkle Reference McCorkle1993; Roberts Reference Roberts1992). Such genuine ambivalence might initially appear more difficult to reconcile with McCullough et al.'s model. Yet, in order to understand the output of any computational system, we must simultaneously analyze the system's informational needs and the information offered by the environment (Gigerenzer et al. Reference Gigerenzer and Todd1999). As McCullough et al. emphasize throughout their article, modern environments do not necessarily deliver the cues that the revenge/forgiveness systems require to operate adaptively. The context within which modern criminal justice intuitions emerge offers a case in point. We evolved to function within small-scale groups and react towards specific exploiters. Modern political debates concerning criminal justice, in contrast, proceed within anonymous mass societies and are general rather than specific in the sense that they are often about how to react towards all criminals. In this context, our counter-exploitation psychology would be activated by the features exhibited by crime, but the lack of ecologically valid cues about the social value of the relevant criminals would make it difficult for this psychology to execute properly and, in particular, to up-regulate either punitive or reparative motivations at the expense of the other. Some evidence for this proposition already exists, as people's difficulties in prioritizing between punitive and reparative goals seem to be restricted to general information-sparse decision-making contexts: When specific cues are directly available, unequivocal criminal justice intuitions emerge (Petersen Reference Petersen2009).
Although the previous applications of the adaptionist framework have emphasized the shared nature of criminal justice intuitions, the widespread existence of conflicting intuitions about the prioritization between punitive and reparative goals should not be taken as evidence against the existence of adapted mechanisms for generating such intuitions. Hence, the model developed by McCullough et al. readily predicts the production of conflicting, changing, and ambivalent intuitions about this particular feature of modern criminal justice.
An adaptationist framework predicts that people's intuitions regarding modern criminal justice stem from psychological mechanisms designed for countering the evolutionarily recurrent equivalent, exploitation (Petersen et al. Reference Petersen, Sell, Tooby, Cosmides and Høgh-Olesen2010). By implication, the model of evolved counter-exploitation mechanisms described by McCullough et al. should predict core features of criminal justice intuitions.
Most of the previous applications of the adaptionist framework to criminal justice focus on people's perceptions of crime seriousness, arguing that they emerge from evolved intuitions about harm (e.g., Robinson et al. Reference Robinson, Kurzban and Jones2007). In making this argument, researchers emphasize the high levels of consensus among individuals and countries on the relative seriousness of different crimes. If widespread agreement is a necessary product of the operations of evolved mechanisms, we would conclude that such mechanisms play little role in the production of the intuitions most relevant to McCullough et al.'s model: that is, intuitions about whether punitive or reparative goals should be given priority in criminal justice. Hence, in criminological research, consensus that people have incoherent or “mushy” criminal justice intuitions has emerged from observations of profound disagreement between individuals, and in different situations about the prioritization of punitive and reparative sanctioning goals (Cullen et al. Reference Cullen, Fisher and Applegate2000; Roberts & Hough Reference Roberts and Hough2002; Roberts & Stalans Reference Roberts and Stalans2004).
McCullough et al.'s model emphasizes (1) the existence of computational systems designed to implement revengeful as well as forgiving strategies, and (2) computations of the social value of the exploiter as the key regulator of the activation of these strategies (see also Petersen et al. Reference Petersen, Sell, Tooby, Cosmides and Høgh-Olesen2010). These two features, I argue, render disagreement between individuals and in different situations an unavoidable outcome of the operation of our evolved counter-exploitation intuitions.
This relates to the fact that the social value of others is both self-specific and target-specific. One target is the self's kin, another is not; in one case, the target and the self share a history of cooperative interactions, in another case they don't; one target is part of the same group as the self, another is not. And so on. Accordingly, different selves will compute the social value of the same target differently, and a single self will compute the social value of different targets differently. In the face of a specific exploiter, some individuals will therefore experience punitive sentiments, while others will experience reparative sentiments.
In modern mass societies, an individual's personal welfare will rarely be affected by how the state sanctions a single criminal, regardless of that criminal's social value. Nevertheless, given the evolutionary importance hereof, our minds should be designed to automatically pick up on ecologically valid cues to the presence of social value (e.g., expressions of remorse, shared ethnic background, lack of criminal record) and activate punitive and reparative sentiments accordingly. In modern criminal justice, different sanctioning types are aligned with these different sentiments to varying degrees. Cost-imposing sentences such as prison are aligned with punitive sentiments, whereas rehabilitation schemes are more aligned with reparative sentiments (e.g., Cullen & Gendreau Reference Cullen and Gendreau2000). Accordingly, different individuals will prioritize different sanctioning types in the face of the same criminal (e.g., depending on the match between the respective ethnic or racial backgrounds of the self and the criminal), and the same individual's priorities will change in the face of different criminals (depending on the specific cues surrounding each case).
This simple observation readily integrates the adaptationist framework and the demonstrations of disagreement within individuals towards different crimes and between individuals towards the same crime. According to this interpretation, lay intuitions are not “mushy,” and disagreement is driven by the existence rather than lack of principled intuitions. If valid, we should find that whether individuals agree or disagree about appropriate sanctions is predictable from whether there is agreement or disagreement with respect to the perceived future social value of the criminal. Some evidence for this link already exists (Burnette et al. Reference Burnette, McCullough, Van Tongeren and Davis2012; Lieberman & Linke Reference Lieberman and Linke2007; Petersen et al. Reference Petersen, Sell, Tooby and Cosmides2012).
Modern criminal justice intuitions have been deemed “mushy” for more than just these kinds of disagreement. In particular contexts, research has also demonstrated how people find it highly difficult to prioritize either reparative or punitive sanctioning goals. Instead, people concurrently express firm support for both (Cullen et al. Reference Cullen, Fisher and Applegate2000; McCorkle Reference McCorkle1993; Roberts Reference Roberts1992). Such genuine ambivalence might initially appear more difficult to reconcile with McCullough et al.'s model. Yet, in order to understand the output of any computational system, we must simultaneously analyze the system's informational needs and the information offered by the environment (Gigerenzer et al. Reference Gigerenzer and Todd1999). As McCullough et al. emphasize throughout their article, modern environments do not necessarily deliver the cues that the revenge/forgiveness systems require to operate adaptively. The context within which modern criminal justice intuitions emerge offers a case in point. We evolved to function within small-scale groups and react towards specific exploiters. Modern political debates concerning criminal justice, in contrast, proceed within anonymous mass societies and are general rather than specific in the sense that they are often about how to react towards all criminals. In this context, our counter-exploitation psychology would be activated by the features exhibited by crime, but the lack of ecologically valid cues about the social value of the relevant criminals would make it difficult for this psychology to execute properly and, in particular, to up-regulate either punitive or reparative motivations at the expense of the other. Some evidence for this proposition already exists, as people's difficulties in prioritizing between punitive and reparative goals seem to be restricted to general information-sparse decision-making contexts: When specific cues are directly available, unequivocal criminal justice intuitions emerge (Petersen Reference Petersen2009).
Although the previous applications of the adaptionist framework have emphasized the shared nature of criminal justice intuitions, the widespread existence of conflicting intuitions about the prioritization between punitive and reparative goals should not be taken as evidence against the existence of adapted mechanisms for generating such intuitions. Hence, the model developed by McCullough et al. readily predicts the production of conflicting, changing, and ambivalent intuitions about this particular feature of modern criminal justice.